This week will be all about slippage: five days and five posts to understand and eliminate slippage.
Don't bother adding contingency to your schedules:
* The amount is made up.
Companies use 20% because that's "the industry standard" even though no one has ever proven this number was any industry's standard. It's simply a number that company executives like.
* No contingency time has ever remained unused.
If it was really contingency it would only be used on projects that need it. Instead, it's time that is always used - so it's not contingency time as much as development time. And if it's development time why not assign it where it's needed?
* It's giving a bad example to the team
It's saying: we know things will go wrong but we can't predict them, we can't prevent them and we can't control them. It's showing the team that being unprepared shouldn't prevent work starting as long as more time is added at the end. They will use the same method to estimate their work: high level guesses with 20% added at the end. Is it any surprise if the team's estimate are inaccurate?
* Contingency time is dangerous
It is usually a lump of time at the end of a schedule so everyone on the team thinks they can use that time for their work. But of course, once you factor in dependencies and the ancillary work (design, testing, integration, review, etc.) only a very small amount of work wil fit in that time.
In addition, team members may think they can count on the contingency and plan their work accordingly
Instead of planning random contingency time, work out the project's risks and unknowns and start by eliminating these. You'll be left with a set of known work that you can then estimate more accurately. It will take time and increase the length of the project - but that's the same whether it's done at the beginning or at the end: do it before work starts and it's called pre-production and it makes your team look professional; do it after work starts and it's called slippage and it makes you team look like fumbling newbies.
Take control of the unknown: give it a name, a shape, a size - and then fight it until it becomes something you know and tame.